IDEAS home Printed from https://ideas.repec.org/a/gam/jeners/v15y2022i14p5085-d861150.html
   My bibliography  Save this article

Physics-Based Method for Generating Fully Synthetic IV Curve Training Datasets for Machine Learning Classification of PV Failures

Author

Listed:
  • Michael W. Hopwood

    (Department of Statistics and Data Science, University of Central Florida, Orlando, FL 32816, USA
    Sandia National Laboratories, Albuquerque, NM 87123, USA)

  • Joshua S. Stein

    (Sandia National Laboratories, Albuquerque, NM 87123, USA)

  • Jennifer L. Braid

    (Sandia National Laboratories, Albuquerque, NM 87123, USA)

  • Hubert P. Seigneur

    (Florida Solar Energy Center, University of Central Florida, Cocoa, FL 32922, USA)

Abstract

Classification machine learning models require high-quality labeled datasets for training. Among the most useful datasets for photovoltaic array fault detection and diagnosis are module or string current-voltage (IV) curves. Unfortunately, such datasets are rarely collected due to the cost of high fidelity monitoring, and the data that is available is generally not ideal, often consisting of unbalanced classes, noisy data due to environmental conditions, and few samples. In this paper, we propose an alternate approach that utilizes physics-based simulations of string-level IV curves as a fully synthetic training corpus that is independent of the test dataset. In our example, the training corpus consists of baseline (no fault), partial soiling, and cell crack system modes. The training corpus is used to train a 1D convolutional neural network (CNN) for failure classification. The approach is validated by comparing the model’s ability to classify failures detected on a real, measured IV curve testing corpus obtained from laboratory and field experiments. Results obtained using a fully synthetic training dataset achieve identical accuracy to those obtained with use of a measured training dataset. When evaluating the measured data’s test split, a 100% accuracy was found both when using simulations or measured data as the training corpus. When evaluating all of the measured data, a 96% accuracy was found when using a fully synthetic training dataset. The use of physics-based modeling results as a training corpus for failure detection and classification has many advantages for implementation as each PV system is configured differently, and it would be nearly impossible to train using labeled measured data.

Suggested Citation

  • Michael W. Hopwood & Joshua S. Stein & Jennifer L. Braid & Hubert P. Seigneur, 2022. "Physics-Based Method for Generating Fully Synthetic IV Curve Training Datasets for Machine Learning Classification of PV Failures," Energies, MDPI, vol. 15(14), pages 1-16, July.
  • Handle: RePEc:gam:jeners:v:15:y:2022:i:14:p:5085-:d:861150
    as

    Download full text from publisher

    File URL: https://www.mdpi.com/1996-1073/15/14/5085/pdf
    Download Restriction: no

    File URL: https://www.mdpi.com/1996-1073/15/14/5085/
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Isidoro Lillo-Bravo & Pablo González-Martínez & Miguel Larrañeta & José Guasumba-Codena, 2018. "Impact of Energy Losses Due to Failures on Photovoltaic Plant Energy Balance," Energies, MDPI, vol. 11(2), pages 1-23, February.
    2. Pillai, Dhanup S. & Rajasekar, N., 2018. "A comprehensive review on protection challenges and fault diagnosis in PV systems," Renewable and Sustainable Energy Reviews, Elsevier, vol. 91(C), pages 18-40.
    3. Chine, W. & Mellit, A. & Lughi, V. & Malek, A. & Sulligoi, G. & Massi Pavan, A., 2016. "A novel fault diagnosis technique for photovoltaic systems based on artificial neural networks," Renewable Energy, Elsevier, vol. 90(C), pages 501-512.
    4. Livera, Andreas & Theristis, Marios & Makrides, George & Georghiou, George E., 2019. "Recent advances in failure diagnosis techniques based on performance data analysis for grid-connected photovoltaic systems," Renewable Energy, Elsevier, vol. 133(C), pages 126-143.
    Full references (including those not matched with items on IDEAS)

    Citations

    Citations are extracted by the CitEc Project, subscribe to its RSS feed for this item.
    as


    Cited by:

    1. Małgorzata Jastrzębska, 2022. "Installation’s Conception in the Field of Renewable Energy Sources for the Needs of the Silesian Botanical Garden," Energies, MDPI, vol. 15(18), pages 1-28, September.

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Li, B. & Delpha, C. & Diallo, D. & Migan-Dubois, A., 2021. "Application of Artificial Neural Networks to photovoltaic fault detection and diagnosis: A review," Renewable and Sustainable Energy Reviews, Elsevier, vol. 138(C).
    2. Nien-Che Yang & Harun Ismail, 2022. "Voting-Based Ensemble Learning Algorithm for Fault Detection in Photovoltaic Systems under Different Weather Conditions," Mathematics, MDPI, vol. 10(2), pages 1-18, January.
    3. Van Gompel, Jonas & Spina, Domenico & Develder, Chris, 2023. "Cost-effective fault diagnosis of nearby photovoltaic systems using graph neural networks," Energy, Elsevier, vol. 266(C).
    4. Qu, Jiaqi & Sun, Qiang & Qian, Zheng & Wei, Lu & Zareipour, Hamidreza, 2024. "Fault diagnosis for PV arrays considering dust impact based on transformed graphical features of characteristic curves and convolutional neural network with CBAM modules," Applied Energy, Elsevier, vol. 355(C).
    5. Van Gompel, Jonas & Spina, Domenico & Develder, Chris, 2022. "Satellite based fault diagnosis of photovoltaic systems using recurrent neural networks," Applied Energy, Elsevier, vol. 305(C).
    6. Belqasem Aljafari & Siva Rama Krishna Madeti & Priya Ranjan Satpathy & Sudhakar Babu Thanikanti & Bamidele Victor Ayodele, 2022. "Automatic Monitoring System for Online Module-Level Fault Detection in Grid-Tied Photovoltaic Plants," Energies, MDPI, vol. 15(20), pages 1-28, October.
    7. Bakdi, Azzeddine & Bounoua, Wahiba & Mekhilef, Saad & Halabi, Laith M., 2019. "Nonparametric Kullback-divergence-PCA for intelligent mismatch detection and power quality monitoring in grid-connected rooftop PV," Energy, Elsevier, vol. 189(C).
    8. Fouzi Harrou & Bilal Taghezouit & Sofiane Khadraoui & Abdelkader Dairi & Ying Sun & Amar Hadj Arab, 2022. "Ensemble Learning Techniques-Based Monitoring Charts for Fault Detection in Photovoltaic Systems," Energies, MDPI, vol. 15(18), pages 1-28, September.
    9. Zixia Yuan & Guojiang Xiong & Xiaofan Fu, 2022. "Artificial Neural Network for Fault Diagnosis of Solar Photovoltaic Systems: A Survey," Energies, MDPI, vol. 15(22), pages 1-18, November.
    10. Chen, Qi & Li, Xinyuan & Zhang, Zhengjia & Zhou, Chao & Guo, Zhiling & Liu, Zhengguang & Zhang, Haoran, 2023. "Remote sensing of photovoltaic scenarios: Techniques, applications and future directions," Applied Energy, Elsevier, vol. 333(C).
    11. Bilal Taghezouit & Fouzi Harrou & Cherif Larbes & Ying Sun & Smail Semaoui & Amar Hadj Arab & Salim Bouchakour, 2022. "Intelligent Monitoring of Photovoltaic Systems via Simplicial Empirical Models and Performance Loss Rate Evaluation under LabVIEW: A Case Study," Energies, MDPI, vol. 15(21), pages 1-30, October.
    12. Fouad Suliman & Fatih Anayi & Michael Packianather, 2024. "Electrical Faults Analysis and Detection in Photovoltaic Arrays Based on Machine Learning Classifiers," Sustainability, MDPI, vol. 16(3), pages 1-29, January.
    13. Antonio Pio Catalano & Ciro Scognamillo & Pierluigi Guerriero & Santolo Daliento & Vincenzo d’Alessandro, 2021. "Using EMPHASIS for the Thermography-Based Fault Detection in Photovoltaic Plants," Energies, MDPI, vol. 14(6), pages 1-19, March.
    14. Selma Tchoketch Kebir & Nawal Cheggaga & Adrian Ilinca & Sabri Boulouma, 2021. "An Efficient Neural Network-Based Method for Diagnosing Faults of PV Array," Sustainability, MDPI, vol. 13(11), pages 1-27, May.
    15. Rouani, Lahcene & Harkat, Mohamed Faouzi & Kouadri, Abdelmalek & Mekhilef, Saad, 2021. "Shading fault detection in a grid-connected PV system using vertices principal component analysis," Renewable Energy, Elsevier, vol. 164(C), pages 1527-1539.
    16. Wang, Haizheng & Zhao, Jian & Sun, Qian & Zhu, Honglu, 2019. "Probability modeling for PV array output interval and its application in fault diagnosis," Energy, Elsevier, vol. 189(C).
    17. Gong, Bin & An, Aimin & Shi, Yaoke & Zhang, Xuemin, 2024. "Fast fault detection method for photovoltaic arrays with adaptive deep multiscale feature enhancement," Applied Energy, Elsevier, vol. 353(PA).
    18. Vorachack Kongphet & Anne Migan-Dubois & Claude Delpha & Jean-Yves Lechenadec & Demba Diallo, 2022. "Low-Cost I–V Tracer for PV Fault Diagnosis Using Single-Diode Model Parameters and I–V Curve Characteristics," Energies, MDPI, vol. 15(15), pages 1-31, July.
    19. Romero-Fiances, Irene & Livera, Andreas & Theristis, Marios & Makrides, George & Stein, Joshua S. & Nofuentes, Gustavo & de la Casa, Juan & Georghiou, George E., 2022. "Impact of duration and missing data on the long-term photovoltaic degradation rate estimation," Renewable Energy, Elsevier, vol. 181(C), pages 738-748.
    20. Krzysztof Barbusiński & Paweł Kwaśnicki & Anna Gronba-Chyła & Agnieszka Generowicz & Józef Ciuła & Bartosz Szeląg & Francesco Fatone & Agnieszka Makara & Zygmunt Kowalski, 2024. "Influence of Environmental Conditions on the Electrical Parameters of Side Connectors in Glass–Glass Photovoltaic Modules," Energies, MDPI, vol. 17(3), pages 1-13, January.

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:gam:jeners:v:15:y:2022:i:14:p:5085-:d:861150. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: MDPI Indexing Manager (email available below). General contact details of provider: https://www.mdpi.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.